1. Plant community succession has been a major area of study over the past century with recent research focusing on the importance of initial colonisers following disturbance. Seed addition can accelerate ecosystem regeneration and is a method commonly used by land managers to restore disturbed lands. However, few studies have examined the effects of seeding treatments on long-term community composition.
2. A study was established in 1984 in the Piceance Basin of northwest Colorado to examine how various revegetation seed mixes affect plant community development following disturbance. The site was surveyed again in 2008 and 2009 to assess long-term community development.
3. Initial seed mix resulted in significant differences in plant community composition after 25 years. Seeding with native and exotic early-seral species resulted in a community with significantly more exotic species and mid-seral shrubs, while seeding with native late-seral species resulted in a community dominated by perennial grasses. Additionally, an unseeded control resulted in a vegetation community dominated by both perennial grasses and mid-seral shrubs, but community composition at the species level was considerably different from that of the seeded treatments. However, the plant community composition of each of the three treatments was significantly different from an adjacent undisturbed reference area, which was dominated by the late-seral shrub, Artemisia tridentata, and perennial grasses.
4. Synthesis and applications. Our results illustrate how initial colonisers (seed mix) can strongly affect subsequent community assemblage after 25 years of development. Restoration ecologists should give considerable thought to the species used in a restoration seed mix to ensure the success of restoration designs and to create the desired community assembly and associated ecosystem services.